- Email: [email protected]
Distribution and pharmacokinetics of radiolabeled monoclonal antibody OC 125 after intravenous and intraperitoneal administration in gynecologic tumors
Distribution and pharmacokinetics of radiolabeled monoclonal antibody OC 125 after intravenous and intraperitoneal administration in gynecologic tumors Hidde J. Haisma, PhD, Karan R. Moseley, MD, Anne Battaile, MS, Thomas C. Griffiths, MD, and Robert C. Knapp, MD Boston, Massachusetts Radiolabeled monoclonal antibodies may be useful for radioimmunotherapy of gynecologic tumors. Iodine 131-labeled F(ab') 2 fragments of a monoclonal antibody, OC 125, with specificity for ovarian carcinoma, were used to study the distribution and pharmacokinetics of this antibody in patients with gynecologic tumors. The radiolabeled antibody was injected intravenously or intraperitoneally into 10 patients suspected of having ovarian cancer. Blood and urine samples were used for pharmacokinetic studies, and biopsy specimens were examined for the uptake of antibody. The serum half-life of the labeled antibody was 30 hours after intravenous administration, with 20% of the injected dose per liter detected at 24 hours. After intraperitoneal injection, the appearance of antibody in serum was slow, with a maximum level of 1.4% of the injected dose per liter at 24 hours. Urinary excretion of the radiolabeled antibody was similar for intravenous and intraperitoneal administration, with approximately 50% of the injected dose excreted after 48 hours. Intraperitoneal administration of the radiolabeled antibody resulted in a higher uptake of antibody in the tumor and a lower uptake of antibody in normal tissues. On the basis of this limited study, intraperitoneal administration of radiolabeled antibody is preferred over intravenous administration for radioimmunotherapy of ovarian cancer. (AM J OBSTET GYNECOL 1988;159:843-8.)
Key words: OC 125, monoclonal antibody, radiotherapy, iodine 131
Ovarian cancer has the highest death rate of all gynecologic cancers. After initial surgery, treatment usually includes chemotherapy and/ or radiation therapy.' The recurrent rate is about 50% for patients with either negative or microscopically positive findings at secondlook operation. 2 After chemotherapy, the 5-year survival rate is 10%.' Since ovarian cancer usually remains within the peritoneal cavity and rarely disseminates to distant body sites, future therapy will probably focus on an intraperitoneal approach! Monoclonal antibodies with specificity for ovarian cancer cells can be used to monitor,' diagnose, 6 and treaf this disease. Knowledge of the pharmacokinetics and distribution of such antibodies is required to plan for clinical applications, especially radiotherapeutic ap-
From the Department of Gynecologic Oncology, Brigham and Women's Hospital, Harvard Medical School, and Dana Farber Cancer Institute. Presented at the International Conference on Monoclonal Antibody Immunoconjugates for Cancer, March 1987, San Diego, California. Received for publication May 16, 1987; revised January 28, 1988; accepted May 19, 1988. Reprint requests: H.]. Haisma, PhD, Department of Oncology, Free University Hospital, de Boelelaan 1117, 1081 HV Amsterdam, The Netherlands.
plications, in which high doses of radioisotopes may be used. After intravenous injection, the amount of antibody reaching the tumor is low, and the antibody uptake in other tissues is significant." In attempts to improve antibody localization in target tumors, radiolabeled antibodies have been injected subcutaneously9 or directly into body cavities. 10 Monoclonal antibody OC 125 binds to approximately 80% of nonmucinous epithelial ovarian cancers." The antigen CA 125 is rapidly shed from the cell surface of ovarian cancer cells and can be detected in the sera of patients with ovarian cancer.' A study was undertaken to compare the distribution and pharmacokinetics of iodine 131-labeled F(ab'h fragments of monoclonal antibody OC 125 after intravenous or intraperitoneal administration in patients diagnosed with or suspected of having ovarian cancer.
Material and methods Monoclonal antibody. OC 125 is a murine monoclonal antibody of the IgG, isotype. The antibody was purified from mouse ascites on a protein A Sepharose column, and F(ab'h fragments were prepared as described. 12 Radiolabeling and administration. OC 125 was iodinated with '"I (New England Nuclear, Boston, Mass.) 843
844
Haisma et al.
October 1988
Am
100
100
-
'U 0 0
Ill
~
(])
_.
~
Q)
Q)
.....
(.)
10
X
LlJ Q)
rJl
5
0
0
Q)
>
(])
en
ell
0
(])
( .)
(])
c
"#.
10
?fi
c.
0
50
1:l
(])
'U
J Obstet Gynecol
5
:::::1
E
1.0
:::::1
(.)
0.5
''' ''
Hours after Injection Fig. 2. Clearance of ~' I-Iabeled OC 125 F(ab') 2 from the urine in patients with gynecologic tumors after intravenous (•) or intraperitoneal (O) injection. 1
60 80 100 Hours after Injection Fig. 1. Clearance of u 1I-labeled OC 125 F(ab'), from the blood in patients with gynecologic tumors after intravenous (e) or intraperitoneal (O) injection. The + indicates blood clearance of radiolabeled antibody in patient 5. second injection.
to a specific activity of approximately 2.5 mCi/mg using the one-vial Iodogen method as described. 1" Immunoreactivity was evaluated 14 with OVCA-433 cells. 1' For intravenous injection the labeled antibody was administered in a peripheral vein. For intraperitoneal injection the antibody was diluted in 500 ml of 0.9% NaCl solution and infused into the abdomen through a 16gauge angiographic catheter over a 15- to 20-minute period. Dispersion of the antibody was followed by -y-camera imaging. Patients received potassium iodide (SSKI, Upsher-Smith Laboratories, Minneapolis, Minn.) 48 hours before and for 4 days after injection of the radiolabeled antibody. Patients. To be eligible for this study, a patient had to have either a pelvic mass suspected of being ovarian cancer or a confirmed tissue diagnosis of ovarian cancer. Written informed consent was obtained. The stage, histopathologic types, and age of the patients are listed in Table I. Pharmacokinetic studies. Blood samples were obtained immediately before, within 15 minutes of injection of the radiolabeled antibody, and subsequently at
various intervals for as long as 2 weeks. Serial urine specimens were also collected from each patient. The proportion of radioactivity associated with protein was determined by precipitation with 10% (w/v) trichloroacetic acid. The serum samples were assayed for CA 125 as described elsewhere. 1" Gel filtration of serum and peritoneal fluid samples was performed with either a Sepharose 6B column or a TSK 3000 highperformance liquid chromatography system, with phosphate-buffered saline solution (140 mmol/L NaCl, 0.15 mmol/L KH,PO, 8 mmol/L, Na,HPO, · 7H,O, pH 7.2) eluant. Optical density and radioactivity were determined to estimate the molecular weight of the radioactive fractions as described. 17 Surgical procedure and sample counting. Surgery was performed 1 to 10 days after injection of the radiolabeled antibody. Tumor biopsy specimens were excised along with fat, muscle, peritoneum, and liver samples. Bone marrow was obtained from the iliac crest. Surgical specimens were washed in phosphate-buffered saline solution and weighed, and )'-emission was counted before the tissues were processed for immunohistochemical and radioimmunometric assay. Immunohistochemical studies and radioimmunometric assay of tissues. Immunohistochemical analysis was performed on fresh-frozen tissue sections by means of the avidin-biotin-peroxidase complex system (Vectastain, Vector Labs, Burlingame, Calif.). Negative and
Intravenous vs. intraperitoneal OC 125 845
Volume 159 Number 4
Table I. Patient characteristics Patient No.
Serum CA 125 (Uiml)
Tumor histology
Papillary serous cystadenocarcinoma Serous cystadenocarcinoma Poorly differentiated serous cystadenocarcinoma Endometrioid carcinoma Poorly differentiated papillary cystadenocarcinoma Papillary adenocarcinoma Endometrial carcinoma Borderline papillary serous cystadenoma Papillary serous adenocarcinoma Papillary serous cystadenocarcinoma
2 3
42 36 54
170 761 I020
4 5
54 56
850 9570
6 7 8
50 72 46
I21 48 71
9 10
56 65
478 26
1
(;rade III
3 3 3
lA,
2 3
III III III
3 0
III III
3 3
IV III III
Table II. Antibody dose and tumor distribution
Patient No.
Dose of antibody (mg)
Dose of iodine (mCi)
Surgery (days after injection)
Uptake of antibody in tumor(% doselgm X J0- 3 )
Immunohistochemical analysis Tumor
I
Peritoneum
Tumor CA 125 (U I JAg)
Intravenous administration I 2 3 4
0.46 0.76 0.77 0.94
0.69 Sa 0.45 b Intraperitoneal administration 6 7 8 9 10
0.84 0.45 0.68 0.37 0.52
1.2 1.9 1.7 2.2 1.8 0.72 0.77 1.0 1.01 0.694 1.85
2.7
+++
+++I*
12.5
0.5
+++
+++/-
6.6
3 I
1.9 2.4
+++ +++
+++I+++I-
8.9 0.3
I
7.4 0.4:j: 5.5
+++ +++
-t NA -t
5.9 <0.01 I53.0
2.6
+
+I-
6 No surgery IO No biopsy obtained
3 I No surgery 2
CA 125 in other tissues (fat and muscle) was
positive control antibodies were included in all reactions. To determine tissue antigen content, samples were solubilized in phosphate-buffered saline solution containing 0.1% sodium dodecyl sulfate and centrifuged to remove insoluble material. CA 125levels were determined in the same manner as for blood samples and expressed as CA 125 units per microgram of protein (Bio Rad, Richmond, Calif.). Results
Pharmacokinetics. OC 125 F(ab') 2 fragments were iodinated and immunoreactivities of at least 80% were obtained. Each patient received a single dose of 13 'I-labeled OC 125 either intravenously or intraperitoneally, except for patient 5 who received two intravenous injections. The patients were closely monitored and showed no toxic or adverse effects. -y-Camera im-
aging during intraperitoneal administration showed good dispersion of the antibody throughout the abdominal cavity. Of the radioactivity in the serum more than 90% was precipitable with trichloroacetic acid, which is in contrast with the radioactivity in the urine, of which less than 10% was precipitable. Clearance of ' 31 I from the circulation was monitored for each patient and is shown in Fig. 1. The mean halflife of 131 I after intravenous injection was 30 hours. The one patient, who received two intravenous injections, showed increased clearance of the antibody at the second injection. In patients receiving a peritoneal injection, the appearance of radioactivity in the blood was slow and reached a maximum 24 hours after injection. After this clearance from the circulation seemed to be at the same rate as in patients who received intravenous injections. Urinary clearance for patients receiving both
846
Haisma. et al.
October 1988 Am J Obstet Gynecol
Table III. Antibody uptake in tissues* % injected dose Igm tissue (times 10 - ') t
IV
Tissue Tumod Fat Muscle Peritoneum Liver Bone marrow Blood
2.3 0.3 0.5 0.9 2.1 !.7 3.3
I
± 0.4§ ± 0.1
± 0.2 ± 0.4 ± 0.3
± 0.4 ± !.5
IP 4.8 05 0.6 2.0 0.5 0.8 !.2
Tumor-to-tissue ratio IV
± 3.2 ± 0.4
± 0.5 ± 0.4 ± 0.2 ± 0.2
± 0.2
7.7 4.6 2.6 1.! 1.4 0.7
I
!P
:L7 H.:i
Vi 10.8 6.4 4.4
IV, Intravenous; IP, intraperitoneal. *For details see Table II. t I to 3 days after injection. :j:Ovarian tumor. §SD.
intravenous and intraperitoneal injections was similar, with approximately 50% of the 131 I excreted at 48 hours (Fig. 2). Preinjection serum CA 125 levels ranged from 26 to 9570 U /mi. Immediately after intravenous injection, serum CA 125 levels dropped precipitously and then increased slowly during monitoring. After intraperitoneal injection, no significant change in CA 125 serum levels was observed. Distribution. The distribution of 131 1-labeled OC 125 F(ab')z monoclonal antibody expressed as a percentage of the injected dose per gram of tissue is shown in Table II. Antibody preferentially accumulated at tumor sites compared with normal tissues. The amount of labeled antibody that localized in the tumor was generally higher after intraperitoneal as opposed to intravenous injection, although the uptake varied from patient to patient (Table III). The amount of antibody that accumulated in the tumor after intravenous injection ranged from 0.5 to 6.1 X I0- 3 % injected dose/ gm of tissue. The amount of antibody in the tumor decreased with later times of surgery. For patients who received the antibody intraperitoneally, the amount of antibody in the tumor ranged from 2.6 to 7.4 X I0- 3 % dose/gm at 1 to 3 days after injection. The uptake of antibody in liver and bone marrow was lower after intraperitoneal injection compared with intravenous injection, whereas uptake in the peritoneum was higher after intraperitoneal injection. Uptake of antibody in fat and muscle for patients injected intraperitoneally showed greater variability. Tumor-to-normal tissue ratios ranged from 0. 7 to 7. 7 for intravenous injections and from 2.6 to 10.8 for intraperitoneal injections. One patient with an endometrial tumor showed low uptake of the radiolabeled antibody in the tumor, which is similar to the uptake in fat or muscle. Immunohistochemical studies and radioimmuno-
assay. All tumor sections stained with OC 125 except for one specimen obtained from a patient with endometrial cancer (patient 7). Other tissues did not stain with OC 125 except for tumor cells present in peritoneum. Staining intensity was expressed in a range of I to 4 in increasing intensity, but no correlation was found between staining intensity and uptake of antibody (Table II). Immunohistochemical analysis failed to detect the radiolabeled antibody in tissue sections despite antibody targeting to tumor as determined by -y-counting. Tissue that did not stain with OC 125, including an endometrial tumor, showed uptake of radiolabeled antibody. Antigen levels in tissues as determined by radioimmunoassay showed neither a correlation with results on immunohistochemical staining nor an accumulation of antibody in the tumor (Table II).
Comment Radioimmunotherapy could be an alternative for the treatment of primary, recurrent, or metastatic ovarian cancer. Knowledge of the pharmacokinetics and distribution of antibodies used in this form of treatment is required to evaluate the feasibility of this approach. Quantitative imaging to calculate radiation doses to various organs can be very inaccurate, especially if high -y-energy isotopes such as 131 I are used, and· tomographic imaging is virtually impossible. 18 Actual antibody uptake of organs can be best estimated by sampling tissue after injection of the radiolabeled antibody. In the present study, a well-characterized antibody with specificity for nonmucinous epithelial ovarian carcinoma was radiolabeled with 131 I and injected intravenously or intraperitoneally into patients with suspect or known ovarian cancer. Pharmacokinetic studies of serum samples revealed a dramatic difference between the distribution in patients injected intravenously and those injected intra-
Volume 159 Number 4
peritoneally. Clearance after intravenous injection was exponential with a mean half-life of 30 hours, which compares with the half-life of 19 hours observed by Hnatowich et al.,'" who used F(ab'h fragments radiolabeled with indium 111. After intraperitoneal i~ection very little radioactivity appeared in the blood. Maximum levels were reached after 24 hours and were about 10 times less than the levels obtained after intravenous injection. One patient received two intravenous injections of radiolabeled antibody. In this patient we observed an increase in blood clearance of the second injection (half-life 1 hour), which was possibly from an antibody response, as was also observed by Pimm et al. 20 Urinary clearance of the injected tracer at this time was similar to tnat of other patients, which indicated that the radiolabeled antibody must have accumulated somewhere in the body, most likely in the liver or spleen. The patient showed no side effects from this second injection, but extra care should be taken when patients receive multiple injections. An antibody response could result in serum sickness or anaphylaxis. This might limit the future use of radio labeled antibody in those patients who show an antibody response. The accumulation of antibody in fat and muscle showed great variability after intraperitoneal injection, which may have resulted from contamination with peritoneal fluid during surgical procedures. The values for uptake of antibody in fat and muscle after intravenous injection (range 0.3 to 0.5 x 10- 3 % injected dose/gm of tissue) correlate well with those reported earlier (0.4 to 0.8 x 10-'% i~ected dose/gm of tissue)." 21 The amount of radiolabeled F(ab') 2 antibody that localized in the liver after intravenous injection (2.1 X 10- 3 % injected dose/ gm of tissue) was slightly higher than that reported by Douillard 20 (0.6 to 0.9 x 10-'% injected dose/ gm of tissue), probably because of later sampling of the tissue in that study. Liver, blood, and bone marrow uptake of antibody was halved after intraperitoneal injection compared with intravenous injection. This is probably from the 10-fold decrease in serum levels observed after intraperitoneal injections compared with intravenous injections. The reduced uptake of antibody in these tissues after intraperitoneal administration will result in dramatically lower radiation doses to these organs, which is important when patients are treated with therapeutic doses of 131 1-labeled antibody. More antibody accumulated in the ovarian tumors after intraperitoneal injection (2.3 to 4.8 x 10 _,% injected dose/ gm tissue) than after intravenous injection, but this difference was not significant (Student t test). The uptake of antibody in the tumor was slightly lower than those reported by others (2.7 to 9.3 X 10- 3 % injected dose/gm tissue)."· 21 We had expected that the amount of antibody that localized in the tumor would be higher after intraperitoneal injection as opposed to
Intravenous vs. intraperitoneal OC 125
847
intravenous injection because the antibody has easier access to the tumor. Uptake of antibody, however, was rather similar. CA 125 levels in ascites may be a 1000 times higher than serum CA 125 levels (unpublished observation), and neutralization of the radiolabeled OC 125 by binding to antigen in the ascites 17 could explain the relative low amount of antibody that localized in the tumor after intraperitoneal injection. Attempts to determine the localization of the antibody in the tumors were unsuccessful. We did not find a correlation between antigen levels in tumors and the uptake of antibody, but tissues that did not express CA 125 antigen, including an endometrial tumor, did not localize the radiolabeled antibody. This indicates that other factors, such as size and location of the tumor, play a role in the localization of the antibody into tumor. The favorable pharmacokinetics of intra peritoneally injected antibody and the slightly higher uptake of antibody in the tumor, with lower liver and bone marrow uptake compared with intravenous injections, make the intraperitoneal route of administration the route of choice for radioimmunotherapy of ovarian cancer.
REFERENCES l. Einhorn N, Nilsson B, Sjovall K. Factors influencing sur-
vival in carcinoma of the ovary. Cancer 1985;55:2019-25. 2. Copeland LJ, Gershenson DM. Ovarian recurrences in patients with no macroscopic tumor at second look laparotomy. Obstet Gynecol 1986;68:873-4. 3. Loui KG, Ozols RF, Myers GE. Longterm results of a cisplatin containing combination chemotherapy regimen for the treatment of advanced ovarian carcinoma. J Clin Oneal 1986;4: 1579-85. 4. Feldman GB, Knapp RC. Lymphatic drainage of the peritoneal cavity and its significance in ovarian cancer. AM J 0BSTET GYNECOL 1974;119:991-4. 5. Bast RC, Klug TL, St. John E, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N EnglJ Med 1983;309:883-7. 6. Pateiski N, Shadier PWD, Czerwenka SK, Hamilton G, Burchel ]. Radioimmunodetection in patients with suspected ovarian cancer. J Nucl Med 1985;26: 1369-76. 7. Epenetos AA, Hooker G, Krausz T, Snook D, Bodmer WF, Taylor-PapadimitriouJ. Antibody guided irradiation of malignant ascites in ovarian cancer. Obstet Gynecol 1986;68:71S-4S. 8. Epenetos AA, Snook D, Durbin H, Johnson PM, TaylorPapadimitriou J. Limitations of radiolabeled monoclonal antibodies for localization of human neoplasms. Cancer Res 1986;46:3183-91. 9. Deland FH, Kim EE, Goldenberg DM. Lymphoscintigraphy with radionuclide-labeled antibodies to carcinoembryonic antigen. Cancer Res 1980;40:2997-3000. 10. Hammersmith Oncology Group and Imperial Cancer Research Fund. Antibody guided irradiation of malignant lesions. Lancet l984;ii:1441-3. 11. Kabawat SE, Bast RC, Bhan AK, Welch WR, Knapp RC, Colvin RB. Tissue distribution of coelomic epithelium related antigen recognized by the monoclonal antibody OC 125. Int J Gynecol Pathol 1983;2:275-85. 12. Johnstone A, Thorpe R. Immunochemistry in practice. London: Blackwell Scientific Publications, 1982:53-5.
Haisma et al.
1:3. Haisma HJ, Hilgers], Zurawski VR. Iodination of monoclonal antibodies for diagnosis and therapy using a convenient one vial method. J Nucl Med 1986;27:1890-5. 14. Lindmo T, Boven E, Luttita F. Determination of the immunoreactive fraction of radiolabeled monoclonal antibodies by linear extrapolation to binding at infini.te antigen excess. J Immunol Methods 1984;72:77-89. 15. Berkowitz RS, Kabawat S, Lazarus H, Colvin RC, Knapp RC, Bast RC. Comparison of rabbit heteroantiserum and murine monoclonal antibody raised against a human epithelial ovarian cancer cell line. AM J OBSTET GY!\ECOL 1983; 146:607-17. 16. Klug TL, Bast RC, Niloff JM, Knapp RC, Zurawski VR. Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA125) associated with human epithelium ovarian carcinoma. Cancer Res 1984;44: I 048. 17. Haisma HJ, Knapp RC, Battaile A, Stradtman EW, Zurawski VR. Antibody-antigen complex formation following injection of mi labeled OC125 monoclonal antibody F(ab') 2 fragments into patients with ovarian carcinoma. Int J Cancer 1987;40:758-62.
October l9HH Am .J Obstet Gynecol
18. Mach JP, Buchegger F. Forni M, eta!. Use of radiolabeled monoclonal anti-CEA for the detection of human carcinomas by external scanning and tomoscintigraphy. Immunol Today 1981;2:239-43. 19. Hnatowich DJ, Griffin TW, Kosciuczyk C, eta!. Pharmacokinetics of an indium-Ill labeled monoclonal antibody in cancer patients. J Nucl Med 1985;26:849-58. 20. Pimm MV, Perkins AC, Armitage NC, Baldwin RW. The characterization of blood-borne radiolabels and the effect of anti-mouse IgG antibodies on localization of radiolabeled monoclonal antibody in cancer patients . .J Nucl Med 1985;26: 1011-23. 21. Larson SM, BrownJP, Wright PW, CarrasquilloJA, Hellstrom I, Hellstrom KE. Imaging of melanoma with I-131 labeled monoclonal antibodies. J Nucl Med 1983;24: 123-9. 22. Douillard JY, Lehur PA, Aillet G, Kremer M, Bianco-A reo A, Chatal JF. Immunohistochemical expression and in vivo tumor uptake of monoclonal antibodies with specificity for tumors of the gastrointestinal tract. Cancer Res 1986;46:4221-4.
Synergistic cytotoxicity between dimethyl sulfoxide and antineoplastic agents against ov·1rian cancer in vitro Rodney F. Pommier, MD,a Eugene A. Woltering, MD,a George Milo, PhD," and William S. Fletcher, MD• Portland, Oregon, and Columbus, Ohio Dimethyl sulfoxide is a well-known differentiating agent that has been shown to inhibit tumor growth in vitro. We hypothesized that antineoplastic agents might show synergistic cytotoxicity when combined with 10% dimethyl sulfoxide. Twenty-four malignant ovarian tumors were removed and used in tests to determine the cytotoxicities of 10% dimethyl sulfOl
Key words: Dimethyl sulfoxide, chemotherapy, ovarian cancer Dimethyl sulfoxide (DMSO) is a polar solvent with low molecular weight and many unique properties. DMSO rapidly penetrates intact skin and potentiates From the Division of Surgical Oncology, Department of Surgery, Oregon Health Sciences University," and the Department of Physiologic Chemistry, Comprehensive Cancer Center, The Ohio State University.' Received for publication January 8, 1988; revised April 15, 1988; accepted May 17, 1988. Reprint requests: Eugene A. Woltering, MD, Division of Surgical Oncology L-223, Oregon Health Sciences Universit_v. 3181 S.W. Sam Jackson Park Road, Portland, OR 97201.
848
the transcutaneous absorption of a variety of lowmolecular-weight ("" 1000 daltons) compounds. Jacob' (1964) was the first to suggest potential therapeutic use of this agent. Subsequently, DMSO was used as a differentiating agent that inhibits tumor growth in vitro. DMSO also increases cell-doubling time in tissue culture, decreases anchorage-independent growth in the donogenic assay, and induces morphologic changes consistent with cell differentiation in a monolayer cell culture. 2 These effects were demonstrated in the use of extremely low concentrations (l% to 2%) of DMSO
Key words: OC 125, monoclonal antibody, radiotherapy, iodine 131
Ovarian cancer has the highest death rate of all gynecologic cancers. After initial surgery, treatment usually includes chemotherapy and/ or radiation therapy.' The recurrent rate is about 50% for patients with either negative or microscopically positive findings at secondlook operation. 2 After chemotherapy, the 5-year survival rate is 10%.' Since ovarian cancer usually remains within the peritoneal cavity and rarely disseminates to distant body sites, future therapy will probably focus on an intraperitoneal approach! Monoclonal antibodies with specificity for ovarian cancer cells can be used to monitor,' diagnose, 6 and treaf this disease. Knowledge of the pharmacokinetics and distribution of such antibodies is required to plan for clinical applications, especially radiotherapeutic ap-
From the Department of Gynecologic Oncology, Brigham and Women's Hospital, Harvard Medical School, and Dana Farber Cancer Institute. Presented at the International Conference on Monoclonal Antibody Immunoconjugates for Cancer, March 1987, San Diego, California. Received for publication May 16, 1987; revised January 28, 1988; accepted May 19, 1988. Reprint requests: H.]. Haisma, PhD, Department of Oncology, Free University Hospital, de Boelelaan 1117, 1081 HV Amsterdam, The Netherlands.
plications, in which high doses of radioisotopes may be used. After intravenous injection, the amount of antibody reaching the tumor is low, and the antibody uptake in other tissues is significant." In attempts to improve antibody localization in target tumors, radiolabeled antibodies have been injected subcutaneously9 or directly into body cavities. 10 Monoclonal antibody OC 125 binds to approximately 80% of nonmucinous epithelial ovarian cancers." The antigen CA 125 is rapidly shed from the cell surface of ovarian cancer cells and can be detected in the sera of patients with ovarian cancer.' A study was undertaken to compare the distribution and pharmacokinetics of iodine 131-labeled F(ab'h fragments of monoclonal antibody OC 125 after intravenous or intraperitoneal administration in patients diagnosed with or suspected of having ovarian cancer.
Material and methods Monoclonal antibody. OC 125 is a murine monoclonal antibody of the IgG, isotype. The antibody was purified from mouse ascites on a protein A Sepharose column, and F(ab'h fragments were prepared as described. 12 Radiolabeling and administration. OC 125 was iodinated with '"I (New England Nuclear, Boston, Mass.) 843
844
Haisma et al.
October 1988
Am
100
100
-
'U 0 0
Ill
~
(])
_.
~
Q)
Q)
.....
(.)
10
X
LlJ Q)
rJl
5
0
0
Q)
>
(])
en
ell
0
(])
( .)
(])
c
"#.
10
?fi
c.
0
50
1:l
(])
'U
J Obstet Gynecol
5
:::::1
E
1.0
:::::1
(.)
0.5
''' ''
Hours after Injection Fig. 2. Clearance of ~' I-Iabeled OC 125 F(ab') 2 from the urine in patients with gynecologic tumors after intravenous (•) or intraperitoneal (O) injection. 1
60 80 100 Hours after Injection Fig. 1. Clearance of u 1I-labeled OC 125 F(ab'), from the blood in patients with gynecologic tumors after intravenous (e) or intraperitoneal (O) injection. The + indicates blood clearance of radiolabeled antibody in patient 5. second injection.
to a specific activity of approximately 2.5 mCi/mg using the one-vial Iodogen method as described. 1" Immunoreactivity was evaluated 14 with OVCA-433 cells. 1' For intravenous injection the labeled antibody was administered in a peripheral vein. For intraperitoneal injection the antibody was diluted in 500 ml of 0.9% NaCl solution and infused into the abdomen through a 16gauge angiographic catheter over a 15- to 20-minute period. Dispersion of the antibody was followed by -y-camera imaging. Patients received potassium iodide (SSKI, Upsher-Smith Laboratories, Minneapolis, Minn.) 48 hours before and for 4 days after injection of the radiolabeled antibody. Patients. To be eligible for this study, a patient had to have either a pelvic mass suspected of being ovarian cancer or a confirmed tissue diagnosis of ovarian cancer. Written informed consent was obtained. The stage, histopathologic types, and age of the patients are listed in Table I. Pharmacokinetic studies. Blood samples were obtained immediately before, within 15 minutes of injection of the radiolabeled antibody, and subsequently at
various intervals for as long as 2 weeks. Serial urine specimens were also collected from each patient. The proportion of radioactivity associated with protein was determined by precipitation with 10% (w/v) trichloroacetic acid. The serum samples were assayed for CA 125 as described elsewhere. 1" Gel filtration of serum and peritoneal fluid samples was performed with either a Sepharose 6B column or a TSK 3000 highperformance liquid chromatography system, with phosphate-buffered saline solution (140 mmol/L NaCl, 0.15 mmol/L KH,PO, 8 mmol/L, Na,HPO, · 7H,O, pH 7.2) eluant. Optical density and radioactivity were determined to estimate the molecular weight of the radioactive fractions as described. 17 Surgical procedure and sample counting. Surgery was performed 1 to 10 days after injection of the radiolabeled antibody. Tumor biopsy specimens were excised along with fat, muscle, peritoneum, and liver samples. Bone marrow was obtained from the iliac crest. Surgical specimens were washed in phosphate-buffered saline solution and weighed, and )'-emission was counted before the tissues were processed for immunohistochemical and radioimmunometric assay. Immunohistochemical studies and radioimmunometric assay of tissues. Immunohistochemical analysis was performed on fresh-frozen tissue sections by means of the avidin-biotin-peroxidase complex system (Vectastain, Vector Labs, Burlingame, Calif.). Negative and
Intravenous vs. intraperitoneal OC 125 845
Volume 159 Number 4
Table I. Patient characteristics Patient No.
Serum CA 125 (Uiml)
Tumor histology
Papillary serous cystadenocarcinoma Serous cystadenocarcinoma Poorly differentiated serous cystadenocarcinoma Endometrioid carcinoma Poorly differentiated papillary cystadenocarcinoma Papillary adenocarcinoma Endometrial carcinoma Borderline papillary serous cystadenoma Papillary serous adenocarcinoma Papillary serous cystadenocarcinoma
2 3
42 36 54
170 761 I020
4 5
54 56
850 9570
6 7 8
50 72 46
I21 48 71
9 10
56 65
478 26
1
(;rade III
3 3 3
lA,
2 3
III III III
3 0
III III
3 3
IV III III
Table II. Antibody dose and tumor distribution
Patient No.
Dose of antibody (mg)
Dose of iodine (mCi)
Surgery (days after injection)
Uptake of antibody in tumor(% doselgm X J0- 3 )
Immunohistochemical analysis Tumor
I
Peritoneum
Tumor CA 125 (U I JAg)
Intravenous administration I 2 3 4
0.46 0.76 0.77 0.94
0.69 Sa 0.45 b Intraperitoneal administration 6 7 8 9 10
0.84 0.45 0.68 0.37 0.52
1.2 1.9 1.7 2.2 1.8 0.72 0.77 1.0 1.01 0.694 1.85
2.7
+++
+++I*
12.5
0.5
+++
+++/-
6.6
3 I
1.9 2.4
+++ +++
+++I+++I-
8.9 0.3
I
7.4 0.4:j: 5.5
+++ +++
-t NA -t
5.9 <0.01 I53.0
2.6
+
+I-
6 No surgery IO No biopsy obtained
3 I No surgery 2
CA 125 in other tissues (fat and muscle) was
positive control antibodies were included in all reactions. To determine tissue antigen content, samples were solubilized in phosphate-buffered saline solution containing 0.1% sodium dodecyl sulfate and centrifuged to remove insoluble material. CA 125levels were determined in the same manner as for blood samples and expressed as CA 125 units per microgram of protein (Bio Rad, Richmond, Calif.). Results
Pharmacokinetics. OC 125 F(ab') 2 fragments were iodinated and immunoreactivities of at least 80% were obtained. Each patient received a single dose of 13 'I-labeled OC 125 either intravenously or intraperitoneally, except for patient 5 who received two intravenous injections. The patients were closely monitored and showed no toxic or adverse effects. -y-Camera im-
aging during intraperitoneal administration showed good dispersion of the antibody throughout the abdominal cavity. Of the radioactivity in the serum more than 90% was precipitable with trichloroacetic acid, which is in contrast with the radioactivity in the urine, of which less than 10% was precipitable. Clearance of ' 31 I from the circulation was monitored for each patient and is shown in Fig. 1. The mean halflife of 131 I after intravenous injection was 30 hours. The one patient, who received two intravenous injections, showed increased clearance of the antibody at the second injection. In patients receiving a peritoneal injection, the appearance of radioactivity in the blood was slow and reached a maximum 24 hours after injection. After this clearance from the circulation seemed to be at the same rate as in patients who received intravenous injections. Urinary clearance for patients receiving both
846
Haisma. et al.
October 1988 Am J Obstet Gynecol
Table III. Antibody uptake in tissues* % injected dose Igm tissue (times 10 - ') t
IV
Tissue Tumod Fat Muscle Peritoneum Liver Bone marrow Blood
2.3 0.3 0.5 0.9 2.1 !.7 3.3
I
± 0.4§ ± 0.1
± 0.2 ± 0.4 ± 0.3
± 0.4 ± !.5
IP 4.8 05 0.6 2.0 0.5 0.8 !.2
Tumor-to-tissue ratio IV
± 3.2 ± 0.4
± 0.5 ± 0.4 ± 0.2 ± 0.2
± 0.2
7.7 4.6 2.6 1.! 1.4 0.7
I
!P
:L7 H.:i
Vi 10.8 6.4 4.4
IV, Intravenous; IP, intraperitoneal. *For details see Table II. t I to 3 days after injection. :j:Ovarian tumor. §SD.
intravenous and intraperitoneal injections was similar, with approximately 50% of the 131 I excreted at 48 hours (Fig. 2). Preinjection serum CA 125 levels ranged from 26 to 9570 U /mi. Immediately after intravenous injection, serum CA 125 levels dropped precipitously and then increased slowly during monitoring. After intraperitoneal injection, no significant change in CA 125 serum levels was observed. Distribution. The distribution of 131 1-labeled OC 125 F(ab')z monoclonal antibody expressed as a percentage of the injected dose per gram of tissue is shown in Table II. Antibody preferentially accumulated at tumor sites compared with normal tissues. The amount of labeled antibody that localized in the tumor was generally higher after intraperitoneal as opposed to intravenous injection, although the uptake varied from patient to patient (Table III). The amount of antibody that accumulated in the tumor after intravenous injection ranged from 0.5 to 6.1 X I0- 3 % injected dose/ gm of tissue. The amount of antibody in the tumor decreased with later times of surgery. For patients who received the antibody intraperitoneally, the amount of antibody in the tumor ranged from 2.6 to 7.4 X I0- 3 % dose/gm at 1 to 3 days after injection. The uptake of antibody in liver and bone marrow was lower after intraperitoneal injection compared with intravenous injection, whereas uptake in the peritoneum was higher after intraperitoneal injection. Uptake of antibody in fat and muscle for patients injected intraperitoneally showed greater variability. Tumor-to-normal tissue ratios ranged from 0. 7 to 7. 7 for intravenous injections and from 2.6 to 10.8 for intraperitoneal injections. One patient with an endometrial tumor showed low uptake of the radiolabeled antibody in the tumor, which is similar to the uptake in fat or muscle. Immunohistochemical studies and radioimmuno-
assay. All tumor sections stained with OC 125 except for one specimen obtained from a patient with endometrial cancer (patient 7). Other tissues did not stain with OC 125 except for tumor cells present in peritoneum. Staining intensity was expressed in a range of I to 4 in increasing intensity, but no correlation was found between staining intensity and uptake of antibody (Table II). Immunohistochemical analysis failed to detect the radiolabeled antibody in tissue sections despite antibody targeting to tumor as determined by -y-counting. Tissue that did not stain with OC 125, including an endometrial tumor, showed uptake of radiolabeled antibody. Antigen levels in tissues as determined by radioimmunoassay showed neither a correlation with results on immunohistochemical staining nor an accumulation of antibody in the tumor (Table II).
Comment Radioimmunotherapy could be an alternative for the treatment of primary, recurrent, or metastatic ovarian cancer. Knowledge of the pharmacokinetics and distribution of antibodies used in this form of treatment is required to evaluate the feasibility of this approach. Quantitative imaging to calculate radiation doses to various organs can be very inaccurate, especially if high -y-energy isotopes such as 131 I are used, and· tomographic imaging is virtually impossible. 18 Actual antibody uptake of organs can be best estimated by sampling tissue after injection of the radiolabeled antibody. In the present study, a well-characterized antibody with specificity for nonmucinous epithelial ovarian carcinoma was radiolabeled with 131 I and injected intravenously or intraperitoneally into patients with suspect or known ovarian cancer. Pharmacokinetic studies of serum samples revealed a dramatic difference between the distribution in patients injected intravenously and those injected intra-
Volume 159 Number 4
peritoneally. Clearance after intravenous injection was exponential with a mean half-life of 30 hours, which compares with the half-life of 19 hours observed by Hnatowich et al.,'" who used F(ab'h fragments radiolabeled with indium 111. After intraperitoneal i~ection very little radioactivity appeared in the blood. Maximum levels were reached after 24 hours and were about 10 times less than the levels obtained after intravenous injection. One patient received two intravenous injections of radiolabeled antibody. In this patient we observed an increase in blood clearance of the second injection (half-life 1 hour), which was possibly from an antibody response, as was also observed by Pimm et al. 20 Urinary clearance of the injected tracer at this time was similar to tnat of other patients, which indicated that the radiolabeled antibody must have accumulated somewhere in the body, most likely in the liver or spleen. The patient showed no side effects from this second injection, but extra care should be taken when patients receive multiple injections. An antibody response could result in serum sickness or anaphylaxis. This might limit the future use of radio labeled antibody in those patients who show an antibody response. The accumulation of antibody in fat and muscle showed great variability after intraperitoneal injection, which may have resulted from contamination with peritoneal fluid during surgical procedures. The values for uptake of antibody in fat and muscle after intravenous injection (range 0.3 to 0.5 x 10- 3 % injected dose/gm of tissue) correlate well with those reported earlier (0.4 to 0.8 x 10-'% i~ected dose/gm of tissue)." 21 The amount of radiolabeled F(ab') 2 antibody that localized in the liver after intravenous injection (2.1 X 10- 3 % injected dose/ gm of tissue) was slightly higher than that reported by Douillard 20 (0.6 to 0.9 x 10-'% injected dose/ gm of tissue), probably because of later sampling of the tissue in that study. Liver, blood, and bone marrow uptake of antibody was halved after intraperitoneal injection compared with intravenous injection. This is probably from the 10-fold decrease in serum levels observed after intraperitoneal injections compared with intravenous injections. The reduced uptake of antibody in these tissues after intraperitoneal administration will result in dramatically lower radiation doses to these organs, which is important when patients are treated with therapeutic doses of 131 1-labeled antibody. More antibody accumulated in the ovarian tumors after intraperitoneal injection (2.3 to 4.8 x 10 _,% injected dose/ gm tissue) than after intravenous injection, but this difference was not significant (Student t test). The uptake of antibody in the tumor was slightly lower than those reported by others (2.7 to 9.3 X 10- 3 % injected dose/gm tissue)."· 21 We had expected that the amount of antibody that localized in the tumor would be higher after intraperitoneal injection as opposed to
Intravenous vs. intraperitoneal OC 125
847
intravenous injection because the antibody has easier access to the tumor. Uptake of antibody, however, was rather similar. CA 125 levels in ascites may be a 1000 times higher than serum CA 125 levels (unpublished observation), and neutralization of the radiolabeled OC 125 by binding to antigen in the ascites 17 could explain the relative low amount of antibody that localized in the tumor after intraperitoneal injection. Attempts to determine the localization of the antibody in the tumors were unsuccessful. We did not find a correlation between antigen levels in tumors and the uptake of antibody, but tissues that did not express CA 125 antigen, including an endometrial tumor, did not localize the radiolabeled antibody. This indicates that other factors, such as size and location of the tumor, play a role in the localization of the antibody into tumor. The favorable pharmacokinetics of intra peritoneally injected antibody and the slightly higher uptake of antibody in the tumor, with lower liver and bone marrow uptake compared with intravenous injections, make the intraperitoneal route of administration the route of choice for radioimmunotherapy of ovarian cancer.
REFERENCES l. Einhorn N, Nilsson B, Sjovall K. Factors influencing sur-
vival in carcinoma of the ovary. Cancer 1985;55:2019-25. 2. Copeland LJ, Gershenson DM. Ovarian recurrences in patients with no macroscopic tumor at second look laparotomy. Obstet Gynecol 1986;68:873-4. 3. Loui KG, Ozols RF, Myers GE. Longterm results of a cisplatin containing combination chemotherapy regimen for the treatment of advanced ovarian carcinoma. J Clin Oneal 1986;4: 1579-85. 4. Feldman GB, Knapp RC. Lymphatic drainage of the peritoneal cavity and its significance in ovarian cancer. AM J 0BSTET GYNECOL 1974;119:991-4. 5. Bast RC, Klug TL, St. John E, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N EnglJ Med 1983;309:883-7. 6. Pateiski N, Shadier PWD, Czerwenka SK, Hamilton G, Burchel ]. Radioimmunodetection in patients with suspected ovarian cancer. J Nucl Med 1985;26: 1369-76. 7. Epenetos AA, Hooker G, Krausz T, Snook D, Bodmer WF, Taylor-PapadimitriouJ. Antibody guided irradiation of malignant ascites in ovarian cancer. Obstet Gynecol 1986;68:71S-4S. 8. Epenetos AA, Snook D, Durbin H, Johnson PM, TaylorPapadimitriou J. Limitations of radiolabeled monoclonal antibodies for localization of human neoplasms. Cancer Res 1986;46:3183-91. 9. Deland FH, Kim EE, Goldenberg DM. Lymphoscintigraphy with radionuclide-labeled antibodies to carcinoembryonic antigen. Cancer Res 1980;40:2997-3000. 10. Hammersmith Oncology Group and Imperial Cancer Research Fund. Antibody guided irradiation of malignant lesions. Lancet l984;ii:1441-3. 11. Kabawat SE, Bast RC, Bhan AK, Welch WR, Knapp RC, Colvin RB. Tissue distribution of coelomic epithelium related antigen recognized by the monoclonal antibody OC 125. Int J Gynecol Pathol 1983;2:275-85. 12. Johnstone A, Thorpe R. Immunochemistry in practice. London: Blackwell Scientific Publications, 1982:53-5.
Haisma et al.
1:3. Haisma HJ, Hilgers], Zurawski VR. Iodination of monoclonal antibodies for diagnosis and therapy using a convenient one vial method. J Nucl Med 1986;27:1890-5. 14. Lindmo T, Boven E, Luttita F. Determination of the immunoreactive fraction of radiolabeled monoclonal antibodies by linear extrapolation to binding at infini.te antigen excess. J Immunol Methods 1984;72:77-89. 15. Berkowitz RS, Kabawat S, Lazarus H, Colvin RC, Knapp RC, Bast RC. Comparison of rabbit heteroantiserum and murine monoclonal antibody raised against a human epithelial ovarian cancer cell line. AM J OBSTET GY!\ECOL 1983; 146:607-17. 16. Klug TL, Bast RC, Niloff JM, Knapp RC, Zurawski VR. Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA125) associated with human epithelium ovarian carcinoma. Cancer Res 1984;44: I 048. 17. Haisma HJ, Knapp RC, Battaile A, Stradtman EW, Zurawski VR. Antibody-antigen complex formation following injection of mi labeled OC125 monoclonal antibody F(ab') 2 fragments into patients with ovarian carcinoma. Int J Cancer 1987;40:758-62.
October l9HH Am .J Obstet Gynecol
18. Mach JP, Buchegger F. Forni M, eta!. Use of radiolabeled monoclonal anti-CEA for the detection of human carcinomas by external scanning and tomoscintigraphy. Immunol Today 1981;2:239-43. 19. Hnatowich DJ, Griffin TW, Kosciuczyk C, eta!. Pharmacokinetics of an indium-Ill labeled monoclonal antibody in cancer patients. J Nucl Med 1985;26:849-58. 20. Pimm MV, Perkins AC, Armitage NC, Baldwin RW. The characterization of blood-borne radiolabels and the effect of anti-mouse IgG antibodies on localization of radiolabeled monoclonal antibody in cancer patients . .J Nucl Med 1985;26: 1011-23. 21. Larson SM, BrownJP, Wright PW, CarrasquilloJA, Hellstrom I, Hellstrom KE. Imaging of melanoma with I-131 labeled monoclonal antibodies. J Nucl Med 1983;24: 123-9. 22. Douillard JY, Lehur PA, Aillet G, Kremer M, Bianco-A reo A, Chatal JF. Immunohistochemical expression and in vivo tumor uptake of monoclonal antibodies with specificity for tumors of the gastrointestinal tract. Cancer Res 1986;46:4221-4.
Synergistic cytotoxicity between dimethyl sulfoxide and antineoplastic agents against ov·1rian cancer in vitro Rodney F. Pommier, MD,a Eugene A. Woltering, MD,a George Milo, PhD," and William S. Fletcher, MD• Portland, Oregon, and Columbus, Ohio Dimethyl sulfoxide is a well-known differentiating agent that has been shown to inhibit tumor growth in vitro. We hypothesized that antineoplastic agents might show synergistic cytotoxicity when combined with 10% dimethyl sulfoxide. Twenty-four malignant ovarian tumors were removed and used in tests to determine the cytotoxicities of 10% dimethyl sulfOl
Key words: Dimethyl sulfoxide, chemotherapy, ovarian cancer Dimethyl sulfoxide (DMSO) is a polar solvent with low molecular weight and many unique properties. DMSO rapidly penetrates intact skin and potentiates From the Division of Surgical Oncology, Department of Surgery, Oregon Health Sciences University," and the Department of Physiologic Chemistry, Comprehensive Cancer Center, The Ohio State University.' Received for publication January 8, 1988; revised April 15, 1988; accepted May 17, 1988. Reprint requests: Eugene A. Woltering, MD, Division of Surgical Oncology L-223, Oregon Health Sciences Universit_v. 3181 S.W. Sam Jackson Park Road, Portland, OR 97201.
848
the transcutaneous absorption of a variety of lowmolecular-weight ("" 1000 daltons) compounds. Jacob' (1964) was the first to suggest potential therapeutic use of this agent. Subsequently, DMSO was used as a differentiating agent that inhibits tumor growth in vitro. DMSO also increases cell-doubling time in tissue culture, decreases anchorage-independent growth in the donogenic assay, and induces morphologic changes consistent with cell differentiation in a monolayer cell culture. 2 These effects were demonstrated in the use of extremely low concentrations (l% to 2%) of DMSO